There's gold in them thar standards!

Megan McArdle (a libertarian economics blogger):

Someone rather more partial to Ron Paul's arguments in favor of the gold standard than I am asks me to write a post outlining my objections to it. All right, here goes.

Money is a mysterious thing. It is a store of value, it is a medium of exchange. It is, in a fiat currency economy, worth only what people think it is worth, and what they think it is worth can be oddly affected by what they think it may be worth in the future, resulting in self-fulfilling feedback loops (at least in the short term). Even in non-fiat currencies, such as the gold standard, the value of the underlying asset can be changed by rising (or shrinking) demand for money. Economists studying this fascinating topic tend to suffer from migraines as they suffer from all the mysterious--hell, nearly mystical--attributes of money.

However, over the last fifty years, economists have settled on some very broad areas of consensus. The first is, as famous libertarian monetary economist Milton Friedman wrote, "inflation is always and everywhere a monetary phenomenon". When the supply of money outstrips the demand, prices rise. And this is by no means limited to fiat currencies; see the great Spanish inflation of the 16th & 17th centuries, thanks to the steady influx of gold from the New World. Or check out the price of basic commodities in mining towns during the Gold Rush, when all anyone had was gold.
The second is that a little bit of inflation is okay--possibly even beneficial, since it helps the economy to overcome the problem of sticky wages when the relative value of labour has fallen. But a lot of inflation is very, very bad. Exhibit A is Zimbabwe; Exhibits B-∞ are every other economy that has had inflation near or above the double-digit mark; the higher the inflation, the worse the economy did. The feeling that the currency will experience an unpredictable amount of inflation dampens the willingness of the citizens to save and invest, which is why so many third-world loans are denominated in dollars.
The third is that deflation is also bad, and at the lower percentage values, often even worse than inflation. This surprises/offends/meets with the frank disbelief of many "sound money" types, who think that, barring local shortage, in an ideal world everything ought to cost the same or less than it did when Grandpa was a boy. (These sorts of opinions are cemented further by the fact that Grandpa, who is often the source of them, is usually living on a fixed income, and therefore feels that he would make out better in a deflationary economy.) The problem is, deflation does rather devastating things to anyone who has debt, since they now have to repay what they borrowed in more expensive dollars. Deflation means that, thanks to the abovementioned sticky wages, the economy has to deal with demand shocks by lowering output. Deflation can result in what's known as a liquidity trap, a concept pioneered by liberal economist John Maynard Keynes and best elucidated by liberal economist Paul Krugman back before he left economics writing to focus on his hatred of George W. Bush. Deflation is what made the Great Depression so memorable. Deflation is so bad that almost everyone agrees that moderate inflation, in the range of 1-2%, is better than risking even a small amount of deflation.
Advocates of a gold standard dispute this. They argue that America experienced a long, slow deflation throughout most of the 19th century, without anyone getting hurt. What they neglect to mention is that people did get hurt, repeatedly, in the period's awful financial contractions. Though we don't have modern economic statistics for the period, it's pretty clear that recessions were longer and deeper than they are now.
This is not only due to the gold standard; the era's primitive financial system and its approach to financial regulation, which often ranged between lighthearted and foolhardy, also played substantial roles. But the gold standard also has to stand up and take a bow. There's a strong correlation, for example, between how long a country hewed to the gold standard, and how badly it suffered from the Great Depression.
The gold standard cannot do what a well-run fiat currency can do, which is tailor the money supply to the economy's demand for money. The supply of gold grows--or not--depending on how much of the stuff is mined. Demand also fluctuates for non-economic reasons; gold has uses besides being money, like industrial components and jewelry.
The lone advantage of a gold standard--and it is a real advantage--is that it prevents governments from inflating the currency. The problem is, this is only moderately true. The government, after all, can always modify its gold standard. Yes, you say, but it will pay a price in the markets, and this is true, but this is the same price it pays when it prints more fiat currency. Such practices do not go unnoticed for long.
As James Hamilton has pointed out, gold-backed currencies, like all money with a fixed exchange rate, are subject to speculative attacks whenever the government's financial position looks weak. Such speculative attacks often require punitive economic measures to fight off, which is one of the reasons that America suffered so nastily from the Great Depression--it raised interest rates in the middle of a recession in order to defend the credibility of its currency.
Also, since devaluations tend to produce sharp changes in the values of currencies, rather than smooth appreciations or declines, the economic dislocations are magnified. Imagine you're a company with a contract denominated in dollars. If the value of the dollar gradually declines, you lose a little, but not too much, since you periodically renew the contract, giving you time to adjust the amounts. If, on the other hand, the devaluation pressure builds up over a period of years, and then all at once the government has to devalue by 20%, you end up badly hurt. You might go out of business. Now multiply that all across the country, and you can see why recessions used to last for years.
In short, you don't get anything out of a gold standard that you didn't bring with you. If your government is a credible steward of the money supply, you don't need it; and if it isn't, it won't be able to stay on it long anyway. (See Argentina's dollar peg). Meanwhile, the limitations on the government's ability to respond to fiscal crises, the necessity of defending against speculative attacks in times of crises, and the possibility of independent changes in the relative price of gold, make your economy more unstable. It's a terrible idea, which is why there are so few economists willing to raise their voices in support of it.